Search Results (1,188)

Satellite biomass products require locally relevant field information, but destructively measured reference data are rare in Mediterranean mountain beech forests. We used 60 felled Fagus sylvatica L. trees sampled in 2023–2024 in the Central Rhodope Mountains, Greece, to develop local stem-carbon models and an exploratory field-to-satellite scale-transfer diagnostic. Smalian stem volumes were converted to dry matter and carbon using a basic wood density of 0.58 Mg m⁻³ and a carbon fraction of 0.47. Log-linear models were evaluated by leave-one-out cross-validation (LOOCV) and bootstrap resampling. Mean stem carbon was 0.165 Mg C tree⁻¹ (95% confidence interval: 0.116–0.221). The diameter at breast height (DBH) plus height model performed best (LOOCV coefficient of determination R² = 0.963, root mean square error RMSE = 0.041 Mg C tree⁻¹, and mean absolute percentage error MAPE = 14.4%). Satellite outputs from the European Space Agency Climate Change Initiative (ESA CCI) Biomass v6 were clipped to the three forest complexes. The Forest Information System for Europe (FISE) threshold masks had no local pixels, so the diagnostic used the European Forest Genetic Resources Programme (EUFORGEN) × WorldCover forest footprint. Its verified mean 2020 living-biomass carbon density was 71.94 Mg C ha⁻¹. With a literature-based stem fraction of 0.65–0.85, this corresponds to 36.25–47.40 Mg stem C ha⁻¹; the central value is 41.83 Mg stem C ha⁻¹, or about 253 sample-mean-equivalent reference stems ha⁻¹. Because the 2020 map predates the 2023–2024 field campaign, these equivalents are exploratory scaling quantities, not evidence of temporal agreement, measured stand density or pixel-level validation. Full article

Forest planning and operational execution still often rely on separate office-based workflows, field inventories, and analogue tree-marking procedures, limiting direct communication between planners and field personnel. Mixed reality (MR) interfaces and forest digital twin concepts may help link planning decisions with field-level visualization and interaction. This study presents an initial proof-of-concept prototype for MR-based virtual tree tagging and field–office communication in forestry. The prototype integrates a Unity GE-based Forest Planner System, a Microsoft HoloLens 2-based Forest Worker System, and a locally hosted MagicOnion server for bidirectional client–server communication. Virtual trees with predefined stem diameters were created in Unity GE, and torus-shaped markers were implemented to support tree selection, color-coded designation, marker deletion, and visualization of predefined diameter-related attributes. Under controlled indoor conditions using a 5G mobile hotspot, both clients connected to the server and exchanged marker-related events. Marker creation, color assignment, deletion, and shared marker-state updates initiated from either client were reproduced in the corresponding system. The prototype demonstrates the functional feasibility of the core communication workflow but remains limited to a simulated environment. Full article

Understanding the upstream-attempt behavior of fish in high-flow environments is important for refining behavioral simulations, optimizing fish-passage design, and protecting aquatic ecosystems. This study investigates Schizothorax oconnori, an endemic fish in the Yarlung Tsangpo River on the Qinghai–Tibet Plateau, using a multi-model analytical framework integrating a generalized linear mixed-effects model, Random Forest, SHapley Additive exPlanations, and GeoDetector. Results showed that remaining endurance and nominal flow velocity were positively associated with fish upstream-attempt initiation. Attempt termination reflected endurance depletion, initial endurance, and hydraulic variability. Model-predicted termination probability increased markedly when (i) cumulative endurance consumption during the current attempt exceeded ~11% or (ii) fish began an attempt with an initial endurance status below ~95%. These model-derived change points represent within-attempt endurance depletion and reduced endurance available at attempt initiation, respectively. Across attempt sequences, the dominant explanatory factors showed an apparent shift from physiological condition during the first attempt to hydraulic variability during subsequent attempts. This sequence-related pattern is consistent with behavioral adjustment based on repeated experience in the flow environment, although cumulative fatigue, stress, individual differences, and experimental procedures may also contribute. Hydraulic-variability metrics along the trajectory showed greater explanatory importance than several instantaneous hydraulic variables, and physiological condition and hydraulic variability together enhanced explanatory power. This study provides a quantitative multi-model framework for analyzing upstream-attempt behavior and offers a scientific basis for refining behavioral simulations and optimizing fish-passage facilities. Full article

Lore Lindu National Park (LLNP), a United Nations Educational, Scientific and Cultural Organization (UNESCO) Biosphere Reserve in Central Sulawesi, Indonesia, is characterized by its significant biodiversity and ecotourism potential, including endemic forest orchids and rich cultural heritage. Despite its ecological importance, the park continues to face deforestation and illegal land use, partly driven by limited community involvement in conservation governance. In this study, we examine the co-creation of governance in community-based forest orchid cultivation and ecotourism initiatives in Karunia Village, where local communities have developed organic orchid cultivation practices since 2004. Drawing on the co-creation governance framework of Christopher Ansell, Eva Sørensen, and Jacob Torfing, we employ a qualitative case study approach based on a literature review, field observations, and in-depth interviews conducted over four months. Our findings reveal that each phase of co-creation is shaped by distinct institutional and socio-political dynamics. Initiation is strongly influenced by local economic pressures and social solidarity, while design encounters regulatory barriers related to licensing and conservation policy. Implementation remains constrained by fragmented coordination among multi-level stakeholders, and systematic evaluation mechanisms are largely absent. We propose the use of a Responsible, Accountable, Consulted, Informed (RACI) Matrix to clarify institutional roles and strengthen collaborative governance arrangements. This article contributes to the literature by expanding the empirical understanding of co-creation practices in the Global South and highlighting the importance of institutional and socio-political dimensions in community-based environmental governance. Full article

Over the past two decades, air-quality prediction has moved from a mainly single-model paradigm toward ensemble systems that make explicit use of diversity across models, observations, and data streams. This review connects developments that are often treated separately: chemical transport model (CTM) ensembles, tree-based and hybrid machine learning ensembles, deep learning architectures, physics-informed neural networks, and distributed approaches such as federated learning. Evidence summarized from recent systematic reviews and coordinated modeling initiatives indicates that, within comparable validation settings, ensembles often outperform individual models for PM_2.5, PM₁₀, O₃, NO₂, CO, and SO₂ across a broad range of spatial scales and standard error metrics, including RMSE, MAE, and correlation. Operational CTM ensembles, such as the Copernicus Atmosphere Monitoring Service (CAMS) European system with eleven regional models, improve both forecast skill and uncertainty characterization for ozone and particulate matter. In data-driven applications, tree-based ensembles (Random Forest, gradient boosting, XGBoost, LightGBM) and hybrid deep architectures (CNN–LSTM models, attention-based multi-branch networks, graph neural networks) now form a core part of the state of the art for AQI (Air Quality Index) and particulate-matter estimation from structured and multi-source data. Reported performance can be very high on well-structured tabular datasets, with $R^2$ values above 0.99 in selected benchmarks and RMSE reductions of 23–45% relative to classical statistical baselines in multi-modal studies; however, these values are not directly interchangeable because pollutant type, prediction horizon, monitoring density, and validation design differ among studies. This review proposes a practical taxonomy of ensemble strategies and uses it to explain why diversity, rather than model count alone, is central to reliable air-quality prediction. Drawing on coordinated European and North American model-evaluation initiatives (AQMEII, HTAP) and on case studies in topographically and meteorologically complex Italian regions (the Po Valley, the Naples metropolitan area, and Campania), we show that effective ensemble design requires a balance among diversity, redundancy, computational feasibility, and interpretability. On the basis of a structured narrative synthesis, the main research gaps concern physics-informed and explainable ensemble frameworks, transferable and adaptive models, standardized benchmarks, severe-pollution-episode forecasting, and scalable distributed architectures. Open questions include how to design compact non-redundant CTM sub-ensembles and how to couple deep learning with chemical-transport physics in next-generation operational systems. Full article

Compression index (Cc) and recompression index (Cur) are essential parameters in one-dimensional consolidation and settlement analysis, yet their direct determination from oedometer testing is time-consuming, costly, and often limited by sparse recompression data. This study develops an interpretable and physically constrained machine-learning framework for the joint prediction of Cc and Cur from four routinely measured index properties: liquid limit (LL), plasticity index (PI), initial void ratio (e), and natural water content (w). A curated subset of 459 natural clay records from the global CLAY/Cc/6/6203 database was used to benchmark single-output and multi-output Random Forest, gradient-boosted tree, and deep neural network models. In addition to conventional random train–test and cross-validation protocols, a leave-one-location-out validation was introduced to evaluate transferability across 81 Country–Location groups. Under the random-split setting, Cc was predicted with moderate-to-good accuracy, with baseline models achieving test R² values of approximately 0.61–0.70 and a geotechnically enriched Random Forest model increasing the test R² to 0.777. Cur was more difficult to predict. Although feature enrichment improved its test R² to 0.507, location-aware validation reduced Cur performance substantially, confirming its stronger dependence on site-specific stress history, fabric, and geological structure. SHAP interpretation identified e and w as the dominant controls on Cc, while Cur exhibited weaker and more diffuse dependence on the available index properties. A physically constrained target transformation based on the bounded ratio of Cur/Cc guaranteed mechanically admissible predictions with Cur < Cc, but did not fully recover the missing information needed for accurate Cur estimation. The proposed constraint is not a governing-equation-based physics-informed model. Rather, it is a mechanically constrained target transformation that preserves the admissible relationship Cur < Cc. The results show that routine index properties can support the useful preliminary prediction of Cc, whereas Cur should be treated as a screening-level estimate unless explicit stress history descriptors are available. Full article

Background: Retinoblastoma (RB) is a rare pediatric malignancy characterized by small sample sizes and low event rates, where conventional association-driven feature selection may lead to unstable models, overadjustment, and limited generalizability. However, existing survival prediction studies lack a careful treatment of feature selection that accounts for underlying causal structure. Objectives: To develop and validate a causality-guided machine learning model for RB survival prediction by jointly incorporating survival time and survival status as outcome variables. Methods: We analyzed 1015 RB patients from the SEER database (1975–2020). A causality-informed feature selection framework was developed to address the challenges of rare-disease data. Specifically, candidate variables were evaluated through a three-step evidence-integration process: (1) univariate Cox proportional hazards (CPH) analysis for initial statistical screening; (2) causal structure learning using the PC algorithm on the variables retained from Step 1 to construct a directed acyclic graph (DAG) and exclude structurally inappropriate variables (colliders or descendants of the outcome); and (3) LASSO-based feature screening performed independently on the full set of candidate variables. The final features were obtained by taking the intersection of the variables retained from Step 2 and Step 3. Survival models were then trained using the selected features, with model comparison performed as a secondary step. Results: The proposed framework consistently identified four structurally and prognostically robust predictors—laterality, “SEER historic stage A”, “RX Summ”, and sequence number—through this evidence-integration process. Compared with conventional approaches, the causality-informed framework reduced the feature set while improving model stability and interpretability. Notably, compared with LASSO-only selection, which retained a larger set of variables, the causality-informed approach yielded a more parsimonious feature set with improved predictive performance, suggesting reduced overfitting in a low-event setting. Survival models trained on this refined feature set demonstrated reliable predictive performance, with the random survival forest achieving the highest discrimination (C-index = 0.739). Importantly, the selected predictors aligned with clinically plausible pathways in the learned DAG, supporting their causal relevance. Conclusions: This study demonstrates that incorporating causal structure into feature selection provides a more reliable and interpretable foundation for survival modeling in retinoblastoma. Rather than focusing on algorithmic comparison alone, our findings highlight that careful, causality-informed feature selection is critical for improving robustness in rare-disease prediction tasks. This framework may serve as a generalizable methodological template for other rare clinical settings prone to spurious associations. Full article

Satellite -derived environmental features can predict county-level life expectancy (LE) across the contiguous United States with a mean absolute error of 1.08 years over two decades, without using any census or sociodemographic inputs. We assembled 61,680 county-year observations across 3084 counties from 2000–2019, integrating features from 11 satellite and gridded data streams. The data streams include the Moderate Resolution Imaging Spectroradiometer (MODIS) land surface temperature and vegetation indices, Sentinel-1 synthetic aperture radar, Sentinel-2 and Landsat optical imagery, the United States Department of Agriculture (USDA) Cropland Data Layer, the European Commission Joint Research Centre (JRC) Global Surface Water layer, the Copernicus Digital Elevation Model, the European Space Agency Climate Change Initiative (ESA CCI) soil moisture record, and the Food and Agriculture Organization (FAO) gridded livestock densities. After a supervised pruning step that removed low-importance variables, a Random Forest regressor was trained and evaluated using 5-fold cross-validation grouped by county. The grouping places all 20 years of each county exclusively in either the training set or the test set, which prevents spatial information leakage between folds. Coefficient of determination, mean absolute error, and root mean squared error are reported as $R^2=0.631\pm 0.013$, MAE $=1.08\pm 0.02$ years, and RMSE $=1.48\pm 0.04$ years. Moran’s I, a measure of residual spatial autocorrelation, is $0.0988$ ($p=0.001$), which supports geographic generalisation. Multimodal fusion reduces unexplained variance by approximately one-third relative to the strongest single-modality baseline (MODIS land surface temperature alone, $R^2=0.442$). TreeSHAP attribution analysis reveals a feature hierarchy in which nighttime land surface temperature features carry roughly $6.16\times$ the cumulative attribution weight of all daytime channels combined. The model response shows a protective inflection near a minimum overnight temperature of about 7.5 °C. Because all input streams are globally available, the framework is architecturally extensible to regions where civil registration and vital statistics systems are incomplete; however, the trained model and its thresholds require recalibration against local mortality data before application outside the contiguous United States. With that caveat, the approach supports satellite-based monitoring of United Nations Sustainable Development Goal (UN SDG) Target 3.9. Full article

In 2019, the United Nations’ General Assembly established 2021 to 2030 as the Decade on Ecosystem Restoration, and ecological restoration should be adopted by the member nations. In 2015, Brazil had already committed to restoring (replanting) twelve million hectares of forests, and this commitment was reaffirmed in the National Plan for the Recovery of Native Vegetation in 2017 and relaunched at COP16 on diversity in 2024. Despite Brazil’s leadership in establishing the Tropical Forests Forever Fund (TFFF) in 2023, which was launched at COP30 in Belem in 2025, the expansion of the agricultural frontier remains the main driver of deforestation in the Amazon/Rain Forest and the Cerrado biomes. This article aims to examine the social and ecological consequences of the capitalist occupation and expansion of the agricultural frontier in the Cerrado. It will also study the counterpoint of the land struggles and initiatives of peasant organizations focused on conservation and restoration as possibilities and perspectives for the social and ecological restoration of the Cerrado landscapes. Based on an interdisciplinary approach, the specialized literature, and official agricultural data, the study shows that, in addition to degrading nature (deforestation, water and soil contamination, and desertification) and threatening the historical ways of life of countryside peoples, the frontier’s expansion blocks possibilities for restoration and hinders initiatives to protect the remaining nature of Brazil’s second-largest biome. On the other hand, resistance to expropriation and appropriation, and struggles for land and territory, have emerged as possibilities for socio-environmental restoration, beyond reforestation and the recovery of destroyed nature, by transforming landscapes, ways of life, and production, and by creating conditions for food sovereignty and sustainability in the countryside. Therefore, agroecological actions by agrarian movements and rural organizations in general, and those of the Movement of Landless Rural Workers (MST) in particular, have become emblematic in opposing agrarian extractivism and unsustainable monocrops imposed upon and disseminated throughout the Brazilian Cerrado. Full article

Artificial intelligence (AI) adoption within logistics organizations remains uneven despite increasing digital transformation initiatives in emerging economies. This study investigates respondent-perceived organizational AI readiness profiles in Thailand’s logistics industry using an integrated analytical framework combining TOE–UTAUT predictors, clustering analysis, supervised machine learning, and explainable artificial intelligence techniques. Data were collected from 520 logistics and supply chain professionals in Thailand using a structured questionnaire. K-means clustering was applied to identify internally derived respondent-perceived AI readiness profiles, while Random Forest, Support Vector Machine (SVM), XGBoost, and LightGBM models were developed to classify readiness-profile membership. A weighted voting ensemble model was additionally employed to assess classification robustness and profile-differentiation stability across multiple learning algorithms. The findings identified three internally derived respondent-perceived AI readiness profiles representing relatively low, moderate, and advanced readiness patterns within the TOE–UTAUT feature space. Among the evaluated models, the SVM classifier achieved the strongest classification performance, obtaining the highest accuracy and AUC values. SHAP analysis indicated that Actual Use, Technological Factors, Facilitating Conditions, and Behavioral Intention exhibited the largest feature-attribution contributions within the readiness-profile classification framework. The study contributes to AI adoption research by integrating clustering-based segmentation, machine-learning classification, and explainable artificial intelligence into a unified readiness-profiling framework. The findings provide practical insights for managers and policymakers seeking to understand respondent-perceived organizational AI readiness patterns and support digital transformation initiatives within logistics professional contexts. Full article

To address the complex factors affecting the productivity of fractured horizontal wells in tight condensate gas reservoirs, as well as the high computational costs and opaque mechanism interpretation associated with traditional numerical simulations, this study proposes and implements a quantitative evaluation method for the main productivity-controlling factors. This method integrates a machine learning surrogate model with the Shapley additive explanations (SHAP) interpretability framework. First, based on 3D geological modeling and fracture propagation simulation, a high-dimensional parameter set encompassing reservoir geology, artificial fractures, and fluid properties was constructed. Subsequently, representative samples were generated through an orthogonal experimental design. On this basis, machine learning algorithms, including Support Vector Machines (SVM), Random Forests (RF), and eXtreme Gradient Boosting (XGBoost), were utilized to construct low-cost, high-precision surrogate models targeting initial productivity and Estimated Ultimate Recovery (EUR). These surrogate models effectively substituted the computationally expensive fully coupled numerical simulations. Furthermore, SHAP values were applied to the trained surrogate models to conduct both global and local interpretability analyses. This approach not only quantifies the magnitude and direction of each input parameter’s contribution to the productivity predictions, but also reveals their non-linear mechanisms and interaction effects. The results indicate that reservoir properties and gas saturation are the fundamental factors determining the productivity of fractured horizontal wells, while fracture conductivity and fracture half-length are the key engineering factors. Furthermore, there exist significant synergistic or antagonistic effects between the geological and engineering parameters. The integrated “parametric modeling–surrogate model construction—SHAP interpretability analysis” workflow established in this study provides a highly efficient, transparent, and physically insightful novel approach for the rapid optimization of fracturing designs and the mechanistic analysis of main productivity-controlling factors in tight condensate gas reservoirs. Full article

To address the challenge of accurately assessing transformer health under multi-source heterogeneous data conditions and to enable precise full-lifecycle lifetime prediction along with interpretable decision-making insights, we propose an ensemble learning framework that integrates multi-model feature selection with Stacking-Optuna optimization. First, we construct a full-lifecycle data framework that integrates a basic data layer with a derived feature layer, resulting in an 89-dimensional feature set. Second, feature importance is evaluated via equal-weighted integration of Random Forest, LightGBM, and XGBoost, and 13 key features are selected using an 80% cumulative importance threshold. On this basis, a Stacking model is constructed using CatBoost, XGBoost, and LightGBM as base learners, with Ridge regression as the meta-learner, while Optuna optimizes the model’s parameters and architecture. Finally, SHAP is employed to perform multi-level interpretability analysis. Experimental results show that the proposed model achieves MAE = 1.3411, RMSE = 1.6899, and R² = 0.9502 on the test set, outperforming the best single model (CatBoost). SHAP analysis further reveals that commissioning year is the feature with the highest SHAP contribution (44.54%), and that feature contributions exhibit a nonlinear pattern, initially decreasing and subsequently increasing with service time. This method achieves high accuracy and interpretability, providing a reference for transformer condition assessment and operational decision-making. Full article

Aiming at the problem that strip surcharge in coastal soft soil foundations causes lateral squeezing and endangers the safety of adjacent existing bridge pile foundations, the traditional Boussinesq elastic theory cannot reflect the true elastoplastic characteristics of the soil and tends to underestimate the actual horizontal additional stress. This paper establishes a two-dimensional plane strain finite element model and, based on the calibration of pure elastic theoretical solutions, carries out extensive comparative analyses under elastoplastic foundation conditions. Through Pearson correlation and random forest sensitivity analyses, it is clarified that the internal friction angle, load ratio, and normalized distance ratio are the core control variables affecting the redistribution of horizontal additional stress, thereby demonstrating the limitations of the influence of elastic modulus and cohesion. The study reveals the nonlinear amplification mechanism of horizontal stress transfer caused by the penetration of the deep plastic zone within the foundation, as well as the physical evolution law of the stress correction factor, which initially exhibits a Gaussian peak enhancement and subsequently decays exponentially with spatial distance. Based on these mechanisms, a combined prediction formula for the horizontal additional stress correction factor is proposed, achieving an R² = 0.903, and a safety evaluation chart for the correction factor is constructed to quantify high-risk areas. The results indicate that when the normalized distance ratio is greater than or equal to 4, the elastoplastic squeezing effect essentially dissipates. The proposed modification method effectively delineates the applicable boundary of the elastic solution and provides a theoretical basis for the bearing capacity calculation and safety control of passively loaded pile foundations in soft soil regions. Full article

Young forests are recognized as important contributors to climate change mitigation due to their high productivity. However, their structural simplicity and transitional nature render them ecologically vulnerable to long-term climatic stress. We explored the long-term responses of young forests to climate change by applying the LANDIS-II forest landscape model coupled with a PnET-based physiological model to simulate 200 years of forest succession and carbon dynamics. Simulations were conducted under three climate scenarios (BAU, RCP45, and RCP85) to evaluate changes in aboveground biomass (AGB), carbon storage, and carbon absorption across elevation gradients. The results revealed that climate change significantly altered successional pathways and carbon capacity, with effects varying with elevation and initial species composition. Predominant species such as Quercus mongolica maintained dominance under the RCP45 and RCP85 scenarios across all elevations, whereas shade-tolerant mid and understory species showed suppressed growth. Sub-alpine species showed prominent declines in AGB, particularly in the RCP85 scenario. These divergent responses increased the spatial heterogeneity of forest productivity and reduced the predictability of forest carbon dynamics over time. Our findings emphasize the uncertainty of predicting forest development and carbon sequestration in young forests under future climatic conditions. They highlight the urgent need to plan forest management strategies incorporating site-specific ecological characteristics, promote successional advancement, and maintain functional stability for effective climate adaptation and mitigation. Full article

Driven by the demand for high-quality housing, fourth-generation residential buildings—known internationally as “Vertical Forest” and in China as “Urban Forest Garden”—have developed rapidly. Initially built in mild southern regions, they have recently expanded to colder northern areas, with over 50 projects underway in provinces such as Shanxi, Hebei, Shaanxi, and Gansu. Several cities have introduced design standards and incentives, and the China Association for Standardization of Engineering Construction has issued the “Design Standards for Urban Forest Garden Housing.” However, in cold regions, where winters are long and cold and summers are short and hot, there is a lack of systematic quantitative research on how balcony design affects building carbon reduction, energy efficiency, and indoor thermal comfort. To address this research gap, this paper poses the following research questions: (1) In fourth-generation residential buildings in cold regions, how do different combinations of balcony orientations affect annual energy consumption and indoor thermal comfort? (2) Which balcony configurations offer the best balance between carbon reduction, energy efficiency, and thermal comfort? Based on statistical analysis of terrace configurations from more than 40 projects, 12 typical configuration models were identified. Using Ladybug and Honeybee tools on the Grasshopper platform, building energy consumption and indoor thermal comfort were simulated. Multi-objective trade-off analysis was performed using the Pareto front method. In this study, indoor thermal comfort was evaluated using the PMV (Predicted Mean Vote) index. PMV is an index proposed by Professor Fanger that comprehensively reflects human thermal sensation, taking into account air temperature, humidity, wind speed, mean radiant temperature, human metabolic rate, and clothing thermal resistance. Its typical range is −3 (cold) to +3 (hot); in this study, the comfort zone was defined as −1 ≤ PMV ≤ 1. Key findings: (1) The southwest + south terrace configuration shows the highest annual energy consumption, exceeding the lowest (northwest + west) by 2.7%, indicating that south-facing terraces are less favorable for carbon reduction. (2) The best thermal comfort is achieved with east, west, and south orientations. Compared to the least comfortable combination (southwest + northwest), the difference in PMV comfort percentage reaches 2.4%. (3) The Pareto front reveals that beyond a certain comfort level, energy consumption increases sharply. The west + south and east + south combinations yield the highest thermal comfort (49.4%) while maintaining relatively low energy consumption (17.98 kWh/m²). Therefore, in cold regions, fourth-generation residential designs should prioritize terrace combinations integrating south-facing and side-facing orientations and avoid pure corner configurations to balance winter solar gain and summer shading. Full article